Abstract
There are several techniques for observing galaxies at high redshift, in the age of reionization : (1) large-z quasars and their Lyman-alpha emission, to pinpoint the nature of the absorption, meadow or forest ;(2) the Lyman breakage technique (LBG), which plots the luminosity function versus z ;(3) Lyman-alpha emitters or LAE ;(4) ALMA tracing of CO [CII] lines ;(5) the use of gravitational lenses, to increase sensitivity at high redshift. While the history of cosmic star formation is well known, at least up to the redshift z = 3, larger redshifts are still uncertain. It does seem that the fall is prolonged after z = 6 - 8, but the uncertainties are great. The number of known quasars between z = 5and 7 has greatly increased (~ 500), but beyond z = 7, they can be counted on the fingers of one hand. Individual quasars are highly ionized, with up to 80 Mpcaround them. This ionized bubble allows photons to escape, and Lyman-alpha emitters are more visible in the vicinity. On the other hand, Lyman-alpha emitters follow large structures and can be observed in groups : more than 2 000 were recently observed with the Subaru telescope, and the shape of the spectra tells us something about their environment (ionized or not). The red-shifted [CII] line allows us to explore massive, star-forming galaxies at the epoch of reionization. Getting started with ALMA was more difficult than expected, due to excitation effects. Simulations need to include other recipes for star formation, feedback and metallicity. It is possible that reionization prevents gas cooling and star formation in dwarf galaxies, which would alleviate the problem of missing satellites in the standard ΛCDM model. The simulations take into account the formation of massive Population III (Pop III) starsonly for the first generation. Black holes can also form very early in rich environments. Observations of quasars at z > 6show that super-massive black holes can form very rapidly, either from Pop III stars or by direct collapse.
